Influence of molybdenum doping on the switching characteristic in silicon oxide-based
resistive switching memory
Yu-Ting Chen, Ting-Chang Chang, Jheng-Jie Huang, Hsueh-Chih Tseng, Po-Chun Yang, Ann-Kuo Chu, Jyun-Bao Yang, Hui-Chun Huang, Der-Shin Gan, Ming-Jinn Tsai, and Simon M. Sze
Citation: Applied Physics Letters 102, 043508 (2013); doi: 10.1063/1.4790277 View online: http://dx.doi.org/10.1063/1.4790277
View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/102/4?ver=pdfcov Published by the AIP Publishing
Articles you may be interested in
Forming-free resistive switching characteristics of 15nm-thick multicomponent oxide Appl. Phys. Lett. 103, 252904 (2013); 10.1063/1.4852059
Conductance quantization in oxygen-anion-migration-based resistive switching memory devices Appl. Phys. Lett. 103, 043510 (2013); 10.1063/1.4816747
Performance and characteristics of double layer porous silicon oxide resistance random access memory Appl. Phys. Lett. 102, 253509 (2013); 10.1063/1.4812474
Understanding the resistive switching characteristics and mechanism in active SiOx-based resistive switching memory
J. Appl. Phys. 112, 123702 (2012); 10.1063/1.4769218
Study of polarity effect in SiOx-based resistive switching memory Appl. Phys. Lett. 101, 052111 (2012); 10.1063/1.4742894
Influence of molybdenum doping on the switching characteristic in silicon
oxide-based resistive switching memory
Yu-Ting Chen,1Ting-Chang Chang,1,2,a)Jheng-Jie Huang,2Hsueh-Chih Tseng,2 Po-Chun Yang,1Ann-Kuo Chu,1Jyun-Bao Yang,1Hui-Chun Huang,3Der-Shin Gan,3 Ming-Jinn Tsai,4and Simon M. Sze2,5
1
Department of Photonics, National Sun Yat-Sen University, Kaohsiung 804, Taiwan
2
Department of Physics and Center for Nanoscience and Nanotechnology, National Sun Yat-Sen University, Kaohsiung 804, Taiwan
3
Department of Materials Science and Engineering, National Sun Yat-Sen University, Kaohsiung 804, Taiwan
4
Electronics and Optoelectronics Research Laboratory, Industrial Technology Research Institute, Chutung, Hsinchu 310, Taiwan
5
Department of Electronics Engineering, National Chiao Tung University, Hsinchu 300, Taiwan
(Received 8 October 2012; accepted 21 January 2013; published online 1 February 2013)
This report compares Mo-doped and undoped SiO2 thin films of a similar thickness as well as
MoOx. The Mo-doped SiO2film exhibited switching behavior after the forming process, unlike the
undoped SiO2film. Through material analyses, a self-assembled layer is observed in the Mo-doped
SiO2 film. Due to the formation of this layer, the thickness required to be broken down is
effectively reduced. Subsequently, the occurrence of the switching behavior in the thinner SiO2
film further confirmed the supposition. A comparison of the two switching behaviors shows that SiO2dominates the switching characteristic of the Mo-doped SiO2.VC 2013 American Institute of
Physics. [http://dx.doi.org/10.1063/1.4790277]
Due to device size scaling requirements, conventional charge storage based memory structures, such as nanocrystal and SONOS, have encountered data storage problems.1–3In recent years, resistive random access memory (ReRAM) has attracted considerable attention as a solution due to its supe-rior advantages.4,5 Resistive switching (RS) characteristics have been discovered in various materials.6–10 Due to their well-developed production processes, well-known material characteristics and wide applications in Si-based thin film, SiO2, or Si-based materials are considered to have the most
potential as candidates for ReRAM. However, the RS behav-ior rarely occurs in SiO2unless metal doping or solid
electro-lyte electrodes are employed.11,12In this paper, this resistive switching issue is investigated in a device utilizing molybdenum-doped SiO2 film. In addition, molybdenum
oxide may transform to molybdenum nanocrystals due to the thermal effect which has been induced by the forming pro-cess.13,14This nanocrystal formation and consequent genera-tion of addigenera-tional oxygen ions per molybdenum atom have been shown to improve the RS performance.8
In this work, standard lithography and reactive ion etch-ing were used to pattern the contact-holes of the cells on the TiN bottom electrode. A 20 nm-thick Mo-doped SiO2 film
(sample A) was deposited by sputtering a MoSi2target in O2
ambient at room temperature. A 200 nm-thick Pt top elec-trode was subsequently stacked. Finally, the devices were completed after a lift-off process. The diagram of the
struc-conductor parameter analyzer. During the forming process, the leakage current of the film increased dramatically and achieved low resistance state (LRS) due to the formation of the conductive filament (CF). When negative voltage was applied, the high resistance state (HRS) was obtained by a gradual increase of resistance during the reset process. Sub-sequently, LRS can be switched back through the set process by applying positive voltage. The morphology and the com-ponents of the films were analyzed by transmission electron microscope (TEM) and X-ray photoelectron spectroscopy (XPS), respectively.
After the forming process by applying positive bias, the RS phenomenon is activated, as shown in the left inset of Fig. 1(b). A stable and repeatable RS behavior is presented in sample A, as shown in Fig. 1(b). The switching mecha-nism is dominated by the oxygen vacancy generation/recom-bination as is demonstrated by the relationship between CF resistance and temperature for LRS, shown in the right inset of Fig.1(a).15,16Through the material analyses of the X-ray photoelectron spectrum, the components of the film can be further confirmed by the Lorenzian-Gaussian fitting. In the photoelectron spectrum of Si 2p3/2, only one peak, that attributed to the SiO2(103 eV), is discovered, as shown in
Fig.1(c).17In the photoelectron spectrum of O1 s, the spec-trum can be separated into two peaks, located at 530.6 eV and 532.4 eV, shown in Fig. 1(d). According to previous research, these two peaks can be assigned as the bonds of
In addition, the switching mechanism of MoOxis due to the
rupture and formation of the metallic filament, as indicated by the proportional relationship between the resistance and the temperature at LRS, shown in the right inset of Fig.
2(a).19–21However, the structure with 20 nm-thick SiO2fails
to be activated during the forming process. The breakdown behavior does not occur even though voltage is applied up to 20 V, as shown in Fig. 2(b). As a result, TEM analysis is employed to determine the difference in the forming process between the Mo-doped SiO2and pure SiO2films. The TEM
image of sample A shown in Fig.2(c)clearly shows a dark region, indicating that a self-assembled layer exists in the
film. The self-assembled layer can be confirmed as MoO3
according to the result of XPS analyses. The thickness of the film required to undergo breakdown might then be reduced due to the existence of the nearly 5 nm-thick self-assembled MoO3layer.
In order to clarify this supposition, a 12-nm-thick SiO2
film is used to cancel out the influence of the self-assembled MoO3 layer (5 nm). After the reduction in the thickness of
the SiO2 film (sample B), a stable RS characteristic is
obtained when the device undergoes the forming process, as shown in Fig. 3(a). The right inset of Fig. 3(a) shows the inversely proportional relationship between the resistance in FIG. 1. (a) Structural diagram and (b) switching behav-ior of the device using Mo-doped SiO2film. XPS
analy-ses of (c) Si2p3/2, and (d) O 1 s. The left and right insets of (b) show the forming characteristic and the temperature-dependence trend of LRS, respectively.
FIG. 2. (a) The unstable switching characteristic of 20 nm-thick MoOx. (b) Activation failure during the
forming process for 23 nm-thick SiO2 film. (c) TEM
image of Mo-doped SiO2film. The left and right insets
of (a) show the switching stability and metallic temper-ature behavior of LRS, respectively.
LRS and ambient temperature, which exhibits the same trend as in sample A. In addition, the RS characteristics of sample A and sample B are compared in Fig.3(b), which shows that
the switching behavior of the device with Mo-doped SiO2
film is nearly the same as that of SiO2, including the
switch-ing voltage and ON/OFF ratio. Therefore, this indicates that switching characteristic of sample A is dominated by SiO2.
However, in addition to affecting the forming process by reducing the thickness required for breakdown, the doped molybdenum also influences RS behavior. The current statis-tics in Fig.4(a) for HRS and LRS extracted at 0.2 V for the two samples show that a lower HRS current occurs in sample A. Since the HRS current is an indicator of the degree of repair for the switching regime of the CF, called switching layer (SL),22 a lower HRS current shows that higher per-formance of the repaired SL is obtained, which might be caused by the existence of more oxygen ions to repair the SL during the reset process.23,24 Next, conduction mechanism analyses are employed and indicate that the conduction mechanism is Schottky emission from 0.48 V to 1 V in HRS for sample A, as shown by the linear fitting between current and root of voltage (I-V1/2). However, Poole-Frenkel emission instead dominates the conduction mechanism along a similar range ( 0.2 V to 0.9 V) for sample B, as shown in Fig.4(b). It is notable that the conduction mechanisms of Fowler–Nordheim tunneling and trap assisted tunneling (TAT) are not supported by fitting analyses, only Poole-Frenkel emission (not shown). These results indicate that the distances of the trap states in the SL, which are attributed to the unrepaired oxygen vacancies, might be too far for charges to transport by tunneling. Therefore, charges trans-port by electric field assisted thermal emission. Furthermore, the conduction mechanism transfer from Poole-Frenkel (sample B) to Schottky emission (sample A) is caused by the improved SL repair during the reset process. MoO3, as
indi-cated by the XPS analysis, can provide three more oxygen ions by breaking all of the bonds of Mo-O during the forming process, with a schematic of the processes shown in FIG. 3. (a) The repeatable switching characteristic of 12 nm-thick SiO2film
and (b) comparison with the switching characteristic of 20 nm-thick Mo-doped SiO2. Left inset of (a) shows stable switching behavior; right inset of
(a) presents temperature behavior of LRS showing that oxygen vacancies dominate the conduction mechanism.
FIG. 4. (a) Statistics for the LRS and HRS in samples A and B. (b) Conduc-tion mechanisms of samples A and B in HRS are, respectively, dominated by Schottky and Poole-Frenkel emissions. (c) Schematic diagrams to illustrate the behaviors after the forming process.
Fig.4(c). As a result, the additional oxygen ions in sample A improve the repaired performance for the SL during the reset process. The higher quality interfacial SL leads to a lower HRS current and the conduction mechanism being domi-nated by Schottky emission.
In summary, although switching behavior can be obtained in the Mo-doped SiO2film, the RS characteristic is not
domi-nated by the MoOx. Because the 20 nm-thick SiO2film failed
in the forming process, a TEM image of the Mo-doped SiO2
film was examined and showed that the self-assembled mo-lybdenum oxide layer reduces the thickness of the film required to be broken down. A thinner SiO2film exhibited a
stable RS characteristic nearly the same as that of the Mo-doped SiO2film. Therefore, the switching behavior of the
Mo-doped SiO2film is in fact dominated by SiO2. The change
in conduction mechanism in HRS between the two devices is due to the additional oxygen ions provided by MoO3.
This work was performed at National Science Council Core Facilities Laboratory for Science and Nano-Technology in Kaohsiung-Pingtung area and was supported by the National Science Council of the Republic of China under Contract NSC 101-2120-M-110-002.
1J. Lu, T. C. Chang, Y. T. Chen, J. J. Huang, P. C. Yang, S. C. Chen, H. C.
Huang, D. S. Gan, N. J. Ho, Y. Shi, and A. K. Chu,Appl. Phys. Lett.96, 262107 (2010).
2T. C. Chang, F. Y. Jian, S. C. Chen, and Y. T. Tsai,Mater. Today14, 608
(2011).
3
S. C. Chen, T. C. Chang, P. T. Liu, Y. C. Wu, P. S. Lin, B. H. Tseng, J. H. Shy, S. M. Sze, C. Y. Chang, and C. H. Lien,IEEE Trans. Electron Device Lett.28, 809 (2007).
4M. Fujimoto, H. Koyama, Y. Nishi, and T. Suzuki,Appl. Phys. Lett.
91, 223504 (2007).
5
H. Shima, F. Takano, H. Muramatsu, H. Akinaga, Y. Tamai, I. H. Inoue, and H. Takagi,Appl. Phys. Lett.93, 113504 (2008).
6D. Choi, D. Lee, H. Sim, M. Chang, and H. Hwang,Appl. Phys. Lett.
88, 082904 (2006).
7
S. C. Chen, T. C. Chang, S. Y. Chen, C. W. Chen, S. C. Chen, S.M. Sze, M. J. Tsai, M. J. Kao, and F. S. Yeh Huang,Solid-State Electron.62, 40 (2011).
8
Y. T. Tsai, T. C. Chang, C. C. Lin, S. C. Chen, C. W. Chen, S. M. Sze, F. S. Yeh(Huang), and T. Y. Tseng,Electrochem. Solid-State Lett.14, H135 (2011).
9
M. C. Chen, T. C. Chang, S. Y. Huang, S. C. Chen, C. W. Hu, C. T. Tsai, and S. M. Sze,Electrochem. Solid-State Lett.13, H191 (2010).
10
J. J. Huang, T. C. Chang, J. B. Yang, S. C. Chen, P. C. Yang, Y. T. Chen, H. C. Tseng, S. M. Sze, A. K. Chu, and M. J. Tsai,IEEE Electron Device Lett.33, 1387 (2012).
11
K. C. Chang, T. M. Tsai, T. C. Chang, Y. E. Syu, C. C. Wang, S. L. Chuang, C. H. Li, D. S. Gan, and S. M. Sze,Appl. Phys. Lett.99, 263501 (2011).
12P. C. Yang, T. C. Chang, S. C. Chen, Y. S. Lin, H. C. Huang, and D. S.
Gan,Electrochem. Solid-State Lett.14, H93 (2011).
13
C. C. Lin, T. C. Chang, C. H. Tu, W. R. Chen, C. W. Hu, S. M. Sze, T. Y. Tseng, S. C. Chen, and J. Y. Lin, Appl. Phys. Lett.93, 222101 (2008).
14
C. H. Tung, K. L. Pey, L. J. Tang, M. K. Radhakrishnan, W. H. Lin, F. Palumbo, and S. Lombardo,Appl. Phys. Lett.83, 2223 (2003).
15M. Liu, Z. Abid, W. Wang, X. He, Q. Liu, and W. Guan,Appl. Phys. Lett.
94, 233106 (2009).
16
H. C. Tseng, T. C. Chang, J. J. Huang, P. C. Yang, Y. T. Chen, F. Y. Jian, S. M. Sze, and M. J. Tsai,Appl. Phys. Lett.99, 132104 (2011).
17J. Yi, X. D. He, Y. Sun, and Y. Li,Appl. Surf. Sci.253, 4361 (2007). 18J. G. Choi, D. Choi, and L. T. Thompson, Appl. Surf. Sci.
108, 103 (1997).
19
K. Jung, H. Seo, Y. Kim, H. Im, J. P. Hong, J. W. Park, and J. K. Lee,
Appl. Phys. Lett.90, 052104 (2007).
20L. Goux, J. G. Lisoni, X. P. Wang, M. Jurczak, and D. J. Wouters,IEEE Trans. Electron Devices56, 2363 (2009).
21
U. Russo, D. Ielmini, C. Cagli, and A. L. Lacaita,IEEE Trans. Electron Devices56, 193 (2009).
22
K. M. Kim, G. H. Kim, S. J. Song, J. Y. Seok, M. H. Lee, J. H. Yoon, and C. S. Hwang,Nanotechnology21, 305203 (2010).
23
S. Yu, Y. Wu, and H.-S. P. Wong, Appl. Phys. Lett. 98, 103514 (2011).
24S. Yu, X. Guan, and H.-S. P. Wong,IEEE Trans. Electron Devices
59, 1183 (2012).
